by parcadeid07ff4b06b62No license3.9K starsListed Oct 8, 2026Updated Oct 8, 2026Repository updated 8 months ago
Async REPL Protocol
Rules for writing code in Agentica's async REPL harness: await Future-returning tools and use one code block per response.
When working with Agentica's async REPL harness for testing.
await for Future-returning toolsCompute AND return in ONE block. Multiple blocks means only first executes.
Source:parcadei/continuous-claude-v3in.claude/skills/archive/async-repl-protocolat commitd07ff4b
License: No license
Content belongs to its original authors. SourceWeft indexes it from a public repository.
parcadei
Displays a dashboard of session token usage, API costs, TLDR compression savings, and hook activity.
parcadei
Guides verification that built infrastructure components are actually invoked in the execution path, not left as dead code.
parcadei
Routes a user's goal to specialist agents and resource levels for research, planning, building, or fixing tasks.
parcadei
Validates a technical plan's technology choices against current best practices and writes a validation handoff.
parcadei
Provides a friendly onboarding overview of an agent's capabilities when users ask what it can do.
parcadei
Produces a token-efficient overview of an unfamiliar codebase using the TLDR stack's tree, structure, call graph and complexity commands.
anthropics
Turns a user's PowerPoint template into a reusable skill that generates branded presentations.
anthropics
Guides creation of Claude Code plugin skills, covering structure, descriptions, progressive disclosure and validation.
anthropics
Guides the structure, manifest, and component layout of Claude Code plugins.
anthropics
Guides creation of Claude Code slash commands, covering structure, YAML frontmatter, arguments and plugin features.
anthropics
Audits CLAUDE.md files in a repository, scores their quality, and applies approved targeted improvements.
vercel-labs
Guides discovery, quality checking and installation of agent skills via the npx skills CLI.
content = await view_file(path) # NOT view_file(path)answer = await ask_memory("...")# GOOD: Single blockcontent = await view_file(path)return any(c.isdigit() for c in content)
# BAD: Split blocks (second block never runs)content = await view_file(path)